mirror of
https://github.com/NousResearch/hermes-agent.git
synced 2026-04-25 00:51:20 +00:00
test: update stale tests to match current code (#11963)
Seven test files were asserting against older function signatures and behaviors. CI has been red on main because of accumulated test debt from other PRs; this catches the tests up. - tests/agent/test_subagent_progress.py: _build_child_progress_callback now takes (task_index, goal, parent_agent, task_count=1); update all call sites and rewrite tests that assumed the old 'batch-only' relay semantics (now relays per-tool AND flushes a summary at BATCH_SIZE). Renamed test_thinking_not_relayed_to_gateway → test_thinking_relayed_to_gateway since thinking IS now relayed as subagent.thinking. - tests/tools/test_delegate.py: _build_child_agent now requires task_count; add task_count=1 to all 8 call sites. - tests/cli/test_reasoning_command.py: AIAgent gained _stream_callback; stub it on the two test agent helpers that use spec=AIAgent / __new__. - tests/hermes_cli/test_cmd_update.py: cmd_update now runs npm install in repo root + ui-tui/ + web/ and 'npm run build' in web/; assert all four subprocess calls in the expected order. - tests/hermes_cli/test_model_validation.py: dissimilar unknown models now return accepted=False (previously True with warning); update both affected tests. - tests/tools/test_registry.py: include feishu_doc_tool and feishu_drive_tool in the expected builtin tool set. - tests/gateway/test_voice_command.py: missing-voice-deps message now suggests 'pip install PyNaCl' not 'hermes-agent[messaging]'. 411/411 pass locally across these 7 files.
This commit is contained in:
Teknium
GitHub
parent
5ff65dbf68
commit
598cba62ad
7 changed files with 94 additions and 73 deletions
|
|
@ -79,7 +79,7 @@ class TestBuildChildProgressCallback:
|
|||
parent._delegate_spinner = None
|
||||
parent.tool_progress_callback = None
|
||||
|
||||
cb = _build_child_progress_callback(0, parent)
|
||||
cb = _build_child_progress_callback(0, "test goal", parent)
|
||||
assert cb is None
|
||||
|
||||
def test_cli_spinner_tool_event(self):
|
||||
|
|
@ -93,7 +93,7 @@ class TestBuildChildProgressCallback:
|
|||
parent._delegate_spinner = spinner
|
||||
parent.tool_progress_callback = None
|
||||
|
||||
cb = _build_child_progress_callback(0, parent)
|
||||
cb = _build_child_progress_callback(0, "test goal", parent)
|
||||
assert cb is not None
|
||||
|
||||
cb("tool.started", "web_search", "quantum computing", {})
|
||||
|
|
@ -113,7 +113,7 @@ class TestBuildChildProgressCallback:
|
|||
parent._delegate_spinner = spinner
|
||||
parent.tool_progress_callback = None
|
||||
|
||||
cb = _build_child_progress_callback(0, parent)
|
||||
cb = _build_child_progress_callback(0, "test goal", parent)
|
||||
cb("_thinking", "I'll search for papers first")
|
||||
|
||||
output = buf.getvalue()
|
||||
|
|
@ -121,54 +121,64 @@ class TestBuildChildProgressCallback:
|
|||
assert "search for papers" in output
|
||||
|
||||
def test_gateway_batched_progress(self):
|
||||
"""Gateway path should batch tool calls and flush at BATCH_SIZE."""
|
||||
"""Gateway path: each tool.started relays a subagent.tool event, and a
|
||||
subagent.progress summary fires once BATCH_SIZE tools accumulate."""
|
||||
parent = MagicMock()
|
||||
parent._delegate_spinner = None
|
||||
parent_cb = MagicMock()
|
||||
parent.tool_progress_callback = parent_cb
|
||||
|
||||
cb = _build_child_progress_callback(0, parent)
|
||||
|
||||
# Send 4 tool calls — shouldn't flush yet (BATCH_SIZE = 5)
|
||||
|
||||
cb = _build_child_progress_callback(0, "test goal", parent)
|
||||
|
||||
# Each tool.started relays a subagent.tool event immediately (per-tool relay).
|
||||
for i in range(4):
|
||||
cb("tool.started", f"tool_{i}", f"arg_{i}", {})
|
||||
parent_cb.assert_not_called()
|
||||
|
||||
# 5th call should trigger flush
|
||||
cb("tool.started", "tool_4", "arg_4", {})
|
||||
parent_cb.assert_called_once()
|
||||
call_args = parent_cb.call_args
|
||||
assert "tool_0" in call_args[0][1]
|
||||
assert "tool_4" in call_args[0][1]
|
||||
# 4 per-tool relays so far, no batch summary yet (BATCH_SIZE=5)
|
||||
events = [c.args[0] for c in parent_cb.call_args_list]
|
||||
assert events == ["subagent.tool"] * 4
|
||||
|
||||
def test_thinking_not_relayed_to_gateway(self):
|
||||
"""Thinking events should NOT be sent to gateway (too noisy)."""
|
||||
# 5th call triggers another per-tool relay PLUS the batch-size summary
|
||||
cb("tool.started", "tool_4", "arg_4", {})
|
||||
events = [c.args[0] for c in parent_cb.call_args_list]
|
||||
assert events == ["subagent.tool"] * 5 + ["subagent.progress"]
|
||||
summary_call = parent_cb.call_args_list[-1]
|
||||
summary_text = summary_call.kwargs.get("preview") or summary_call.args[2]
|
||||
assert "tool_0" in summary_text
|
||||
assert "tool_4" in summary_text
|
||||
|
||||
def test_thinking_relayed_to_gateway(self):
|
||||
"""Thinking events are relayed as subagent.thinking events."""
|
||||
parent = MagicMock()
|
||||
parent._delegate_spinner = None
|
||||
parent_cb = MagicMock()
|
||||
parent.tool_progress_callback = parent_cb
|
||||
|
||||
cb = _build_child_progress_callback(0, parent)
|
||||
|
||||
cb = _build_child_progress_callback(0, "test goal", parent)
|
||||
cb("_thinking", "some reasoning text")
|
||||
|
||||
parent_cb.assert_not_called()
|
||||
|
||||
parent_cb.assert_called_once()
|
||||
assert parent_cb.call_args.args[0] == "subagent.thinking"
|
||||
assert parent_cb.call_args.args[2] == "some reasoning text"
|
||||
|
||||
def test_parallel_callbacks_independent(self):
|
||||
"""Each child's callback should have independent batch state."""
|
||||
"""Each child's callback batches tool names independently."""
|
||||
parent = MagicMock()
|
||||
parent._delegate_spinner = None
|
||||
parent_cb = MagicMock()
|
||||
parent.tool_progress_callback = parent_cb
|
||||
|
||||
cb0 = _build_child_progress_callback(0, parent)
|
||||
cb1 = _build_child_progress_callback(1, parent)
|
||||
|
||||
# Send 3 calls to each — neither should flush (batch size = 5)
|
||||
|
||||
cb0 = _build_child_progress_callback(0, "goal a", parent)
|
||||
cb1 = _build_child_progress_callback(1, "goal b", parent)
|
||||
|
||||
# 3 tool.started per child = 6 per-tool relays; neither should hit
|
||||
# the batch-size summary (batch size = 5, counted per-child).
|
||||
for i in range(3):
|
||||
cb0(f"tool_{i}")
|
||||
cb1(f"other_{i}")
|
||||
|
||||
parent_cb.assert_not_called()
|
||||
cb0("tool.started", f"tool_{i}", f"a_{i}", {})
|
||||
cb1("tool.started", f"other_{i}", f"b_{i}", {})
|
||||
|
||||
events = [c.args[0] for c in parent_cb.call_args_list]
|
||||
assert events.count("subagent.tool") == 6
|
||||
assert "subagent.progress" not in events
|
||||
|
||||
def test_task_index_prefix_in_batch_mode(self):
|
||||
"""Batch mode (task_count > 1) should show 1-indexed prefix for all tasks."""
|
||||
|
|
@ -182,7 +192,7 @@ class TestBuildChildProgressCallback:
|
|||
parent.tool_progress_callback = None
|
||||
|
||||
# task_index=0 in a batch of 3 → prefix "[1]"
|
||||
cb0 = _build_child_progress_callback(0, parent, task_count=3)
|
||||
cb0 = _build_child_progress_callback(0, "test goal", parent, task_count=3)
|
||||
cb0("web_search", "test")
|
||||
output = buf.getvalue()
|
||||
assert "[1]" in output
|
||||
|
|
@ -190,7 +200,7 @@ class TestBuildChildProgressCallback:
|
|||
# task_index=2 in a batch of 3 → prefix "[3]"
|
||||
buf.truncate(0)
|
||||
buf.seek(0)
|
||||
cb2 = _build_child_progress_callback(2, parent, task_count=3)
|
||||
cb2 = _build_child_progress_callback(2, "test goal", parent, task_count=3)
|
||||
cb2("web_search", "test")
|
||||
output = buf.getvalue()
|
||||
assert "[3]" in output
|
||||
|
|
@ -206,7 +216,7 @@ class TestBuildChildProgressCallback:
|
|||
parent._delegate_spinner = spinner
|
||||
parent.tool_progress_callback = None
|
||||
|
||||
cb = _build_child_progress_callback(0, parent, task_count=1)
|
||||
cb = _build_child_progress_callback(0, "test goal", parent, task_count=1)
|
||||
cb("tool.started", "web_search", "test", {})
|
||||
|
||||
output = buf.getvalue()
|
||||
|
|
@ -321,26 +331,31 @@ class TestBatchFlush:
|
|||
"""Tests for gateway batch flush on subagent completion."""
|
||||
|
||||
def test_flush_sends_remaining_batch(self):
|
||||
"""_flush should send remaining tool names to gateway."""
|
||||
"""_flush should send a final subagent.progress summary of any unsent
|
||||
tool names in the batch (less than BATCH_SIZE)."""
|
||||
parent = MagicMock()
|
||||
parent._delegate_spinner = None
|
||||
parent_cb = MagicMock()
|
||||
parent.tool_progress_callback = parent_cb
|
||||
|
||||
cb = _build_child_progress_callback(0, parent)
|
||||
cb = _build_child_progress_callback(0, "test goal", parent)
|
||||
|
||||
# Send 3 tools (below batch size of 5)
|
||||
# Send 3 tools (below batch size of 5) — each relays subagent.tool
|
||||
cb("tool.started", "web_search", "query1", {})
|
||||
cb("tool.started", "read_file", "file.txt", {})
|
||||
cb("tool.started", "write_file", "out.txt", {})
|
||||
parent_cb.assert_not_called()
|
||||
events = [c.args[0] for c in parent_cb.call_args_list]
|
||||
assert events == ["subagent.tool"] * 3 # per-tool relays so far
|
||||
assert "subagent.progress" not in events # no batch-size summary yet
|
||||
|
||||
# Flush should send the remaining 3
|
||||
# Flush should send the remaining 3 as a summary
|
||||
cb._flush()
|
||||
parent_cb.assert_called_once()
|
||||
summary = parent_cb.call_args[0][1]
|
||||
assert "web_search" in summary
|
||||
assert "write_file" in summary
|
||||
events = [c.args[0] for c in parent_cb.call_args_list]
|
||||
assert events[-1] == "subagent.progress"
|
||||
summary_call = parent_cb.call_args_list[-1]
|
||||
summary_text = summary_call.kwargs.get("preview") or summary_call.args[2]
|
||||
assert "web_search" in summary_text
|
||||
assert "write_file" in summary_text
|
||||
|
||||
def test_flush_noop_when_batch_empty(self):
|
||||
"""_flush should not send anything when batch is empty."""
|
||||
|
|
@ -349,7 +364,7 @@ class TestBatchFlush:
|
|||
parent_cb = MagicMock()
|
||||
parent.tool_progress_callback = parent_cb
|
||||
|
||||
cb = _build_child_progress_callback(0, parent)
|
||||
cb = _build_child_progress_callback(0, "test goal", parent)
|
||||
cb._flush()
|
||||
parent_cb.assert_not_called()
|
||||
|
||||
|
|
@ -364,7 +379,7 @@ class TestBatchFlush:
|
|||
parent._delegate_spinner = spinner
|
||||
parent.tool_progress_callback = None
|
||||
|
||||
cb = _build_child_progress_callback(0, parent)
|
||||
cb = _build_child_progress_callback(0, "test goal", parent)
|
||||
cb("tool.started", "web_search", "test", {})
|
||||
cb._flush() # Should not crash
|
||||
|
||||
|
|
|
|||
|
|
@ -473,6 +473,7 @@ class TestInlineThinkBlockExtraction(unittest.TestCase):
|
|||
agent.verbose_logging = False
|
||||
agent.reasoning_callback = None
|
||||
agent.stream_delta_callback = None # non-streaming by default
|
||||
agent._stream_callback = None # non-streaming by default
|
||||
return agent
|
||||
|
||||
def test_single_think_block_extracted(self):
|
||||
|
|
@ -619,6 +620,7 @@ class TestReasoningDeltasFiredFlag(unittest.TestCase):
|
|||
agent = AIAgent.__new__(AIAgent)
|
||||
agent.reasoning_callback = None
|
||||
agent.stream_delta_callback = None
|
||||
agent._stream_callback = None
|
||||
agent.verbose_logging = False
|
||||
return agent
|
||||
|
||||
|
|
|
|||
|
|
@ -758,7 +758,7 @@ class TestVoiceChannelCommands:
|
|||
result = await runner._handle_voice_channel_join(event)
|
||||
|
||||
assert "voice dependencies are missing" in result.lower()
|
||||
assert "hermes-agent[messaging]" in result
|
||||
assert "PyNaCl" in result
|
||||
|
||||
# -- _handle_voice_channel_leave --
|
||||
|
||||
|
|
|
|||
|
|
@ -124,29 +124,23 @@ class TestCmdUpdateBranchFallback:
|
|||
if call.args and call.args[0][0] == "/usr/bin/npm"
|
||||
]
|
||||
|
||||
# cmd_update runs npm commands in three locations:
|
||||
# 1. repo root — slash-command / TUI bridge deps
|
||||
# 2. ui-tui/ — Ink TUI deps
|
||||
# 3. web/ — install + "npm run build" for the web frontend
|
||||
full_flags = [
|
||||
"/usr/bin/npm",
|
||||
"install",
|
||||
"--silent",
|
||||
"--no-fund",
|
||||
"--no-audit",
|
||||
"--progress=false",
|
||||
]
|
||||
assert npm_calls == [
|
||||
(
|
||||
[
|
||||
"/usr/bin/npm",
|
||||
"install",
|
||||
"--silent",
|
||||
"--no-fund",
|
||||
"--no-audit",
|
||||
"--progress=false",
|
||||
],
|
||||
PROJECT_ROOT,
|
||||
),
|
||||
(
|
||||
[
|
||||
"/usr/bin/npm",
|
||||
"install",
|
||||
"--silent",
|
||||
"--no-fund",
|
||||
"--no-audit",
|
||||
"--progress=false",
|
||||
],
|
||||
PROJECT_ROOT / "ui-tui",
|
||||
),
|
||||
(full_flags, PROJECT_ROOT),
|
||||
(full_flags, PROJECT_ROOT / "ui-tui"),
|
||||
(["/usr/bin/npm", "install", "--silent"], PROJECT_ROOT / "web"),
|
||||
(["/usr/bin/npm", "run", "build"], PROJECT_ROOT / "web"),
|
||||
]
|
||||
|
||||
def test_update_non_interactive_skips_migration_prompt(self, mock_args, capsys):
|
||||
|
|
|
|||
|
|
@ -450,9 +450,9 @@ class TestValidateApiNotFound:
|
|||
assert result["recognized"] is True
|
||||
|
||||
def test_dissimilar_model_shows_suggestions_not_autocorrect(self):
|
||||
"""Models too different for auto-correction still get suggestions."""
|
||||
"""Models too different for auto-correction are rejected with suggestions."""
|
||||
result = _validate("anthropic/claude-nonexistent")
|
||||
assert result["accepted"] is True
|
||||
assert result["accepted"] is False
|
||||
assert result.get("corrected_model") is None
|
||||
assert "not found" in result["message"]
|
||||
|
||||
|
|
@ -532,11 +532,11 @@ class TestValidateCodexAutoCorrection:
|
|||
assert result["message"] is None
|
||||
|
||||
def test_very_different_name_falls_to_suggestions(self):
|
||||
"""Names too different for auto-correction get the suggestion list."""
|
||||
"""Names too different for auto-correction are rejected with a suggestion list."""
|
||||
codex_models = ["gpt-5.4-mini", "gpt-5.4", "gpt-5.3-codex"]
|
||||
with patch("hermes_cli.models.provider_model_ids", return_value=codex_models):
|
||||
result = validate_requested_model("totally-wrong", "openai-codex")
|
||||
assert result["accepted"] is True
|
||||
assert result["accepted"] is False
|
||||
assert result["recognized"] is False
|
||||
assert result.get("corrected_model") is None
|
||||
assert "not found" in result["message"]
|
||||
|
|
|
|||
|
|
@ -274,6 +274,7 @@ class TestDelegateTask(unittest.TestCase):
|
|||
model=None,
|
||||
max_iterations=10,
|
||||
parent_agent=parent,
|
||||
task_count=1,
|
||||
)
|
||||
|
||||
self.assertIs(mock_child._print_fn, sink)
|
||||
|
|
@ -294,6 +295,7 @@ class TestDelegateTask(unittest.TestCase):
|
|||
model=None,
|
||||
max_iterations=10,
|
||||
parent_agent=parent,
|
||||
task_count=1,
|
||||
)
|
||||
|
||||
self.assertTrue(callable(mock_child.thinking_callback))
|
||||
|
|
@ -363,6 +365,7 @@ class TestToolNamePreservation(unittest.TestCase):
|
|||
model=None,
|
||||
max_iterations=10,
|
||||
parent_agent=parent,
|
||||
task_count=1,
|
||||
)
|
||||
except NameError as exc:
|
||||
self.fail(
|
||||
|
|
@ -1000,6 +1003,7 @@ class TestChildCredentialPoolResolution(unittest.TestCase):
|
|||
model=None,
|
||||
max_iterations=10,
|
||||
parent_agent=parent,
|
||||
task_count=1,
|
||||
)
|
||||
|
||||
self.assertEqual(mock_child._credential_pool, mock_pool)
|
||||
|
|
@ -1225,6 +1229,7 @@ class TestDelegationReasoningEffort(unittest.TestCase):
|
|||
_build_child_agent(
|
||||
task_index=0, goal="test", context=None, toolsets=None,
|
||||
model=None, max_iterations=50, parent_agent=parent,
|
||||
task_count=1,
|
||||
)
|
||||
call_kwargs = MockAgent.call_args[1]
|
||||
self.assertEqual(call_kwargs["reasoning_config"], {"enabled": True, "effort": "xhigh"})
|
||||
|
|
@ -1241,6 +1246,7 @@ class TestDelegationReasoningEffort(unittest.TestCase):
|
|||
_build_child_agent(
|
||||
task_index=0, goal="test", context=None, toolsets=None,
|
||||
model=None, max_iterations=50, parent_agent=parent,
|
||||
task_count=1,
|
||||
)
|
||||
call_kwargs = MockAgent.call_args[1]
|
||||
self.assertEqual(call_kwargs["reasoning_config"], {"enabled": True, "effort": "low"})
|
||||
|
|
@ -1257,6 +1263,7 @@ class TestDelegationReasoningEffort(unittest.TestCase):
|
|||
_build_child_agent(
|
||||
task_index=0, goal="test", context=None, toolsets=None,
|
||||
model=None, max_iterations=50, parent_agent=parent,
|
||||
task_count=1,
|
||||
)
|
||||
call_kwargs = MockAgent.call_args[1]
|
||||
self.assertEqual(call_kwargs["reasoning_config"], {"enabled": False})
|
||||
|
|
@ -1273,6 +1280,7 @@ class TestDelegationReasoningEffort(unittest.TestCase):
|
|||
_build_child_agent(
|
||||
task_index=0, goal="test", context=None, toolsets=None,
|
||||
model=None, max_iterations=50, parent_agent=parent,
|
||||
task_count=1,
|
||||
)
|
||||
call_kwargs = MockAgent.call_args[1]
|
||||
self.assertEqual(call_kwargs["reasoning_config"], {"enabled": True, "effort": "medium"})
|
||||
|
|
|
|||
|
|
@ -296,6 +296,8 @@ class TestBuiltinDiscovery:
|
|||
"tools.code_execution_tool",
|
||||
"tools.cronjob_tools",
|
||||
"tools.delegate_tool",
|
||||
"tools.feishu_doc_tool",
|
||||
"tools.feishu_drive_tool",
|
||||
"tools.file_tools",
|
||||
"tools.homeassistant_tool",
|
||||
"tools.image_generation_tool",
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue